fbpx

πŸš€ We’re excited to share insights from our latest webinar “Winning Q1: Maximizing Post-Holiday Subscription App Growth” with Singular and Revenue Cat! 🌟 

We dove into strategies for leveraging the dip in CPMs post-Christmas to fuel subscription app growth. πŸ“‰βž‘οΈπŸ“ˆ Key takeaways include the importance of scaling marketing efforts in January, leveraging low CPMs for higher conversion rates, and maintaining user motivation to combat the mid-January motivation drop. πŸ‹οΈβ€β™‚οΈπŸ’‘

We also explored trends across various app categories, revealing insights on consumer behavior and effective marketing channels. πŸ“ŠπŸ” From the resurgence of Black Hat ASO tactics to the creative use of hybrid monetization models, we covered it all. πŸŽ©πŸ’°

Special thanks to our superstar panelists, John Koetsier VP of Insights at Singular, Ekaterina Gamsriegler, Head of Growth and Marketing at MIMO, David Barnard, Growth Advocate at RevenueCat and Virendra Shekhawat, Head of Mobile Marketing at Rocketship HQ.for their invaluable insights and strategies for thriving in Q1. 🌟





ABOUT THE PANELISTS: John Koetsier | Ekaterina Gamsriegler | David Barnard | Virendra Shekhawat

ABOUT ROCKETSHIP HQ: Website | LinkedIn  | Twitter | YouTube


KEY HIGHLIGHTS

πŸ‘Ÿ Leverage the post-holiday dip in CPMs to scale marketing efforts efficiently.

πŸ‡πŸ½ Utilize January’s low CPMs and high conversion rates for growth.

🎀 Adapt strategies to maintain user motivation and reduce churn.

🎳 Explore various channels, including TikTok and SDK networks, for diversification.

🎯 Embrace hybrid monetization models to enhance revenue streams.

🎻 Utilize AI and generative tools for efficient creative production and testing.

πŸš€ Monitor and adapt to changing CPM trends and user behavior post-Christmas.

β›± Experiment with non-digital marketing channels for localized targeting.

🧭 Prepare for privacy changes and the adoption of SKAN 4.0 and Google’s Privacy Sandbox.

πŸ”§ Utilize blended metrics and hybrid measurement for more accurate performance analysis.

FULL TRANSCRIPT BELOW

Shamanth 

Welcome to the webinar that we’re hosting with our friends at Singular and Revenue Cat on Winning Q1 – How to Maximize Your Post Holiday Subscription App Growth. 

We have with us a superstar panel and I’m excited to introduce to you.

Please join me in welcoming John Koetsier VP of Insights at Singular, Ekaterina Gamsriegler, Head of Growth and Marketing at MIMO, David Barnard, Growth Advocate at RevenueCat and Virendra Shekhawat, Head of Mobile Marketing at Rocketship HQ. I’m your host Shamanth Rao, CEO at Rocketship HQ. 

We can jump into the webinar. This is about what’s happening post-Christmas, post-holidays. What are you seeing with the CPM trends post-Christmas? And what impact are you seeing on subscription apps strategies for Q1?

Katya, if you want to take this. 

Ekaterina

Well I think that this trend has been consistent for us, for the education category app for a while. And I’ve seen similar trends for health and fitness apps, the CPMs go down. For us, the first week of January was with very low CPMs, they dropped almost twice, and starting about the seventeenth of January, they started getting back on track.

But I think what a lot of personal development apps might also see during the season is the much higher natural intent to opt in for a trial, to purchase because of the New Year resolutions that a lot of users have. So for us, this is a period to scale up as much as we can, just because even if the CPMs would stay stable we would have benefited greatly from this because of the higher conversion rates.

That’s why typically we’ll launch the channels and I believe other personal development apps do as well, which they might not be running regularly as well as scaling as much as possible, the ones that we are running. The only thing that I would say has to be adapted with the strategy is to be sure that you help users to stay motivated till at least the end of the month ideally for a couple of months ahead because as we all know, and might also personally experience with the new year resolutions, the motivation might go down somewhere middle of the month and be completely gone by the end of the month.

That’s why all the best practices should be in place when it comes to grace periods or helping users to, prevent involuntary churn by reminding users of the progress they’ve made with either CRM or some pop-ups on the product. I think it makes sense to give them as much of a motivational boost as you can to make sure that their moneyhas not been wasted and these subscribers stay with you till the end of the month or longer. 

Shamanth 

There’s the new year, new you and there’s resolutions. Both these come along with the lowest CPMs as well. Those are interesting dynamics and Viren, what are you seeing post-Christmas?

Virendra 

I think very similar to what Ekatrina shared. With some of the apps that we have been managing and more specifically if we speak about the education space we are seeing a very similar effect where there is a decrease in CPMs post-Christmas because holiday campaigns are passed now, there’s a decrease in competition.

The major advertisers are pulling back their spending and as a result, the CPMs are going down. So while we did see earlier that the CPM did go down by like 0.5x or 0.8x, right after mid-January, we are seeing an increase in CPM slightly but not alarming because there’s an increase in conversion rate, which is what Ekaterina also shared that a lot of users are into apps where they are motivated to test out new products and new services.

And then I think with a decrease in CPMs, what we are seeing is there is an increase in budgets because most of the organizations are trying to get the most of decreased CPMs. Of what we are seeing currently we are having an increase in budgets across clients and portfolios. But then again, there’s a huge influx of new clients as well.

But this is a very good time to increase spending and test out new channels because I think it’s the right time to take advantage of it.

Shamanth 

John, going over to you, how are you seeing trends by genre or sub-verticals based on all of the things that you’re seeing? 

John

As we’ve heard, post-Christmas is always a little tough. Everyone’s tired. You spent all your money. You have to catch up on work, and all those emails, and everybody else is kind of like that too.

You have to get back into real life. So Q1’s often a dip from Q4, but there are opportunities. We’re now in the last week of January so we’re seeing things starting to revive. I took a bit of a deep dive on CTR click-through rate, just because it’s indicative of top funnel interest. And just noted some categories and genres that, stood out to me.

So I saw some super low CTR games, categories, casinos, adventure, and sports strategy. They’re 2% below the CTR range. Competitive and tough, with lots of ads in there and not tons of click-through. Some of the categories that are high right now in gaming are racing and music.

So that was interesting. We’re seeing like 11 to 12% click-through there. Of course, that’s the top funnel. What happens down below, what happens in terms of people converting and subscribing is another matter. 

In terms of apps, some of the ones that I saw were super low books, reference, music, audio, productivity, food and drink, sub 1 percent click-through rate. And some of the high click-through rate apps photo and video and social were 4 percent and up. So not as high as the high-end games, but interesting nevertheless. 

Shamanth: 

Interesting, and you’re right about that. The top funnel metrics often are a good leading indicator of what happens down the funnel. 

To switch gears to smaller independent apps that are not as reliant on paid ads. David, you guys work with a number of these smaller independent apps. What are some of the changes and trends that you’re seeing lately?

David 

I wouldn’t say this is a trend per se, but I did want to point out that Black Hat ASO is still a thing. I talked with Steve Young on the SubClub podcast about this back in the summer.

And no matter what time you’re trying to improve your ASO, understanding those black hat techniques that your competitors might be implementing, is always super important. And then anytime there are things like the β€œnew year, new you”, and stuff like that, there will be developers out there using black hat tactics to capitalize on whatever’s currently trending.

So if you’re trying to catch a trend or you’re trying to optimize around specific events or seasons or whatever, just be cognizant of the fact that competitors are probably doing stuff that you shouldn’t do. And you’re going to get kicked off the app store for doing so. 

There are a few other things I’ve seen micro-influencers and affiliates kind of being more creative with advertising for these smaller apps – people who don’t have a 50,000 budget a month, to train the Facebook algorithm. 

I was just talking to a developer recently who once they crossed that threshold and went from that 20,000 spend to like a 50, 60,000 spend, it made a huge difference. Well, that’s impractical for so many apps.

And so seeing a lot of folks getting more and more creative and just kind of getting back to the basics of marketing –  get attention for your app. Well, there’s a lot of different ways to get attention. Building out communities, working with micro-influencers and influencers who already have those communities or are plugged into those communities, I think there are a lot of really creative ways to get attention 

Search ads are still kind of the go-to if you’re just getting started. Those can be competitive. It can be hard to get profitable bids on some of the more competitive keywords, but you can get creative in search ads as well.

John: 

Shamanth, if I could just tag on what David said there, it’s super interesting that you mentioned communities.

I just interviewed the CEO of AllTrails. AllTrails was the 2023 iPhone app of the year. There is one of these apps out of the millions on the app store. And he talked about community and he said it’s his second biggest asset after their internal team and their people. They’re huge on subscriptions.

They make a ton of money on subscriptions. And without a strong community, it doesn’t happen. That’s good for continuity, for engagement, for attention. It’s also good for acquisition. 

Shamanth: 

David, to your earlier point, you’re right that training the Facebook algorithm takes 20, 30 grand a month. That’s one reason why we also have those numbers as our minimum for a lot of our engagements. But to your point, you don’t have to necessarily go the Facebook route. That’s the right route to go if you’re looking to scale, but if you’re smaller, there are certainly alternate ways to grow and stay sustainable.

Katya what have you seen with smaller independent apps? 

Ekaterina: 

Just from my mentoring experience, I also see that of course in recent years become so much harder for them to go to paid social media channels for iOS, simply because of the privacy thresholds you have to cross off the learnings you’re trying to make.

And if we also think about Meta and TikTok, there also is creative fatigue, and you have to regularly produce the new catchy thumbstops and creatives, which is also quite a bit of effort and investment. So just like others, I do see them opting in for Apple search ads as they go at least checking the waters kind of channel.

However, what’s interesting is that of course it’s not the cheapest channel out there and the CPI does not always match the average revenue per user that you’re getting, which makes it still tricky to grow and scale there. And another, which I believe was not the case or would not have been the case two years ago, I see them using their social media channels, especially TikTok is big, just trying to grow through some viral efforts and to grow through there.

I think it has slightly changed in recent years. They have to be creative now. 

Shamanth: 

Certainly, I have seen a lot of independent apps grow a lot through organic TikTok. And that remains a channel that can quite blow up. So you’re right. 

Staying on the topic of channels, Viren, what are you seeing in terms of the channel mix changes lately?

Virendra: 

It goes in sync with the first question on that to ask concerning CPMs. Clients more specifically are more inclined to test out new channels, like TikTok. TikTok was in the news for like last two to three years. People are spending quite a decent amount, but then Facebook is the biggest channel most of the time whenever we have to drive UA. 

But then more recently, everyone is more open to testing out TikTok and then SDK networks like Unity, AppLab, and ironSource, because even now they rely somewhat on probabilistic matching, so that does mean that they have a higher potential for success unlike Facebook, where there’s a big blocker of Privacy from Apple, these SDKs have more chance to succeed.

And that is why a lot of our clients internally, I think, are more open to testing out these channels, even though there is a certain blocker concerning spending, but If the spending is not an issue and if they’re open to scale, I think these channels deserve a higher chance of success. And what is currently happening is with platforms like TikTok and Meta, since subscription ads are very much consumer-oriented.

The UGC set of creatives that we have been testing out earlier, were very specific to TikTok, but now we are seeing a lot of success even on meta. Even in UGC I mean there are so many user motivations. There are so many types of creators based on consumer personas that we can test out and I think that’s what is driving the performance quite a lot on all of the channels that we have been testing more recently So in terms of channel mix, I think people are more open to testing out now unlike before.

Shamanth: 

On SDK networks, we’re seeing a lot more uptake lately, UGC as a creative format, we’re seeing a lot of uptake lately, which I think is very different from before. 

Viren, to build on what you said with the probabilistic attribution. We’re seeing Facebook do something similar with AEM. Google’s been doing firebase. So people are doing stuff that is not SKAN-linked.

Katya, what are you seeing concerning channel mix? 

Ekaterina: 

I also agree with what Viren said it’s very interesting how it’s changing when it comes to the marketing mix. Apart from testing out and being more open to experimenting with different Ad networks, et cetera I think what almost everybody has become very open to, is to also launching the channels, which are typically delayed response channels.

So not as direct return on ad spend driven and obsessed with as we used to be which would be influencer partnerships. Probably YouTube some advertising at the top of the funnel which you still, want to measure. 

We are now way more open to testing things because we no longer expect deterministic attribution with 95 percent certainty.

So I think this is quite an interesting switch. Apart from this, these blurred lines between organic and paid, social media channels, should not even go to advertise on TikTok unless you have grown your organic presence there significantly because a lot of users are checking out your profile, and this is quite an indicator for them before engaging with the ads.

Things like this are also quite interesting. I think also, of course, in the last year a lot of us have tested the web-to-app channels, which are new auctions, new audiences new targeting possibilities, which have also been quite big in 2023. 

David: 

I want to jump in real quick and share a strategy I talked to The CEO of Ladder on the podcast about deterministic advertising. The hack they kind of found around that, and a lot of apps are experimenting with this now, but Ladder has seen just a ton of success in pushing people from the TikTok ads specifically to a web quiz.

And when you push to the web, you do get deterministic attribution. I’ve heard from so many developers who struggle to get the web to work. But if you can get the web to work, it can be incredible because you do get that deterministic attribution. 

And then what they did, which I thought was especially interesting is that their quiz is structured in a way that they understand pretty quickly into the quiz, how likely that person is to become a subscriber and be kind of their ideal customer profile. And so, by the end of the quiz, in minutes, they’re able to send that feedback back to the TikTok algorithm and close that loop quickly and deterministically.

And they’ve seen a ton of success with that. It’s on the Sub Club podcast. It was probably four episodes back if you want to listen to more details about that. But yeah, if you can get the web to work, it can be powerful in this time when driving people directly to the app store can be challenging on measurement.

Shamanth: 

100%. We’ve seen the web to be a huge opportunity. There’s also a high-effort, high-reward play. Not everyone we know has made it work, but those that have, have done very, very well indeed.

We talked about several channels, several interesting non-Meta, non-Google channels. Are you seeing advertisers use non-social or non-digital channels lately? And what forces, if any, are you seeing precipitate these changes? David, if you want to take that.

David: 

I think on the web, Shamanth was talking about how it’s really hard to make the web work and it’s a ton of effort. Ladder, which I was talking about earlier is a fitness app. I think the apps where it works best and the strategies that work best for web subscriptions is when you can drive a ton of intent to the web. 

So if you have low intent if you don’t have a really clear value prop, you’re going to have a really hard time making the web work. If you’re trying to get the web work, you reverse engineer that. 

What’s that value prop? That’s going to click in a way that builds enough intent that we can convert those to the web. And then maybe some of the ads you’re sending to the web and some you’re sending to the app store and It doesn’t have to be one or the other.

But I think that’s one of the keys to getting the web to work. 

Shamanth 

David, what are you seeing in terms of any shift if at all to non-digital non-social channels? 

David 

I have not seen a ton. I did talk to one developer maybe 18 months ago who had a very specific kind of more localized value prop and they blanketed a city with like bus ads, radio ads, and things like that, you would think, how would that work for digital?

But what they found was that it was cheap. This goes back to what I was saying earlier and can apply at any level. It’s not just when you don’t have the money to spend on Meta that you should be creative. Any app at any scale should be looking for creative ways to get attention.

There’s a great example of a developer who saw an opportunity for very low-cost ways to get attention. In local radio ads and stuff. That’s probably not going to work for most apps if not all apps. I see that kind of a trend of just looking for opportunity and then taking advantage of it.

Typically we in the digital space are always looking for those opportunities digitally, but they do exist in billboards, from what I’ve heard, prices of billboards have gone down. 

Now, is that going to work? Probably not, but to be thinking that way, is helpful for a marketing team to not always assume that any creativity has to be executed digitally.

Shamanth: 

We had Lisa from Fishbrain, on our podcast who also talked about how offline partnerships drove a ton of their growth. And what was interesting was that it was very attributable. They had QR codes to track exactly who came from which partner. So it certainly solves one of the attribution problems. 

David, to just build on what you said, I think one reason why this can work for local apps is also that algorithms don’t like local products. They don’t like small audiences. And every time we’ve done a local product with Meta ads, The CPMs go through the roof.

And this is one case where I think local ads can work very well. 

John to switch over to you, what are you seeing in terms of any trends towards non-digital or non-social channels?

John: 

Super interesting what David was talking about with the radio. And of course, you got to know your demo. If your demo is listening to the radio or if they’re on Apple Music or Spotify, that’s a different ball of wax.

So if your demo’s there, great. I looked at this from a bit of a broader perspective, but what I saw, and I’m building out some of the data now for Singular’s 2024 ROI Insights report. Nobody wants to hear this, but the bigger are getting bigger on the digital side. Google, Meta, TikTok, and Apple search ads. 

I am also seeing some non-traditional sources going up – custom SMS. Something that was kind of a big deal like a decade ago and then kind of disappeared and nobody ever heard about it. And all of a sudden you’re starting to see little interesting bits and pieces there. 

Mobile web we just talked about that. A lot cheaper in a lot of cases to get ads on mobile web versus in-app ads. And there is some measurability and trackability even though third-party cookies are going away and privacy sandbox is coming there too. 

CTV – we’re seeing more action there as well. Some other interesting ones that are niche, they’re not huge volume, but I thought were interesting, email. Email still works. And I’m seeing some people reinvest in there and build up more. There are a ton of newsletters going out these days and they have very targeted niche audiences. So you find a good sub-stack newsletter. You find a good other type of newsletter.

And you put an ad or a sponsorship in there, that can be big. Referral, custom integrations, I’ve seen that increase. No store direct install on Android, I’m seeing more of that happening. And maybe, of course, that’ll happen on iOS. Perhaps in the EU in the next couple of years or so with the Digital Markets Act.

Single tap by Digital Turbine, we’ve seen some interesting things there. Also, lock screen ads, platforms like Glance by Inmobi are interesting things and affiliate networks. I’m doing the data right now for the 2024 singular ROI index. There are some very interesting changes happening.

Shamanth: 

One of the things I love about these webinars is the sheer variety of ideas that come up, and, also realizing how little I do know. So thank you for sharing all of that. 

To switch gears just a little bit, we talked earlier in preparation for this webinar about something that’s emerging and new with a lot of subscription apps, which is hybrid monetization models.

Talk to us about what. Do you mean by hybrid monetization models? What some of these changes are what’s driving these changes? David, if you can take that 

David: 

Hybrid monetization at a high level is just multiple income streams from your digital user base. Fish brand is a great example who’s been doing this for a long time? They have a marketplace, they have ads, they have sponsorships, they have a lot of things going on. And then I was struck today. I got a research note on Netflix’s most recent quarterly results. And, it just dawned on me. They’re a perfect example of hybrid monetization becoming a big deal in the subscription world. You know, they’ve been subscription-exclusive for a very long time. And they’ve started to bump up against growth problems. A lot of subscription apps start to saturate. And you see this with CPM starting to go up and cost per customer acquisition costs go up as an app matures because you’ve picked a low-hanging fruit of willing subscribers.

Netflix being one of the biggest consumer subscription companies in the world, they’ve picked most of the low-hanging fruit on increasing subscriber growth. So what are they turning to? Hybrid monetization. First, they’re turning to raising prices, but that’s not always a good option unless you’re Netflix and even they get some pushback.

What we saw them do last year was introduce an ad tier. So now, they’re not doing free ads. They’re doing paid with ads, but that’s a form of hybrid monetization. And then they’re also and this isn’t hybrid monetization specific, but they’re also diversifying the value prop with their gaming, which is increasing the value to users and taking that bundle approach where people use multiple parts of your product, they’re less likely to churn because there’s more value and more touch points with your brand.

 And then specifically on hybrid monetization, a Sub Club podcast just came out today where I spoke with Surfline, and it’s the title of the podcast is How to Succeed with Freemium and Hybrid Monetization. And they’re doing some great stuff. And I’ve heard from other apps like Onyx that are doing it as well. And that’s taking a kind of holistic approach to your ads and sponsorship. 

So on Surfline, they do ads for the free users. There are just so many people who use Surfline. And we talked about this on the podcast but not everyone is going to subscribe. Like they’re just not going to, and slowly, Surfline is picking away at more and more of those and they’re adding more value to entice those users. But they do show ads to the non-subscribed users and for Surfline with the webcams, that’s video ads, which can be more profitable than just your typical banner ad.

But when they sign on a new sponsor, they think about it much more holistically. So they do ads from those sponsors that show to the free users, but then they also do deals. And this is kind of a win-win where there’s a deal for subscribers and it just gives the brand access to subscribers who otherwise wouldn’t see ads.

The fact that they’re subscribers means they’re probably higher-value people for the brand to reach anyway. And it doesn’t feel like an ad when you say, hey, get 30 percent off your next surfboard or whatever. Surfline gets to promote that as a value prop for the subscription. Hey, if you’re a subscriber to Surfline, you get these discounts that in some cases end up paying for the cost of the subscription.

The last thing I wanted to say on this point is that you don’t have to be on X or Surfline to do this kind of thing. Way back in 2011 I had a fuel economy and maintenance tracking app. Fram auto parts reached out to me and said β€œHey, we’re interested in doing something.”

For them to sponsor my app, if somebody changes their oil at 3000 miles instead of 6000 miles, it’s a huge increase in profit for them over time if people change their oil filter more frequently. And so way back then with my little tiny indie app, I signed a brand deal to sponsor this massive company inside of my app.

Would I have been able to land that reaching out? I don’t know. But it’s one of those things that even as a smaller app and anywhere on the spectrum, there’s potential to do these kinds of deals. If you can find those kinds of win-wins for sponsorships, for ads, for whatever different kind of monetization you can use.

And one of the important things I think a lot of freemium apps miss is understanding the value that those free users are contributing. If you’re free users, aren’t seeing ads, they’re not contributing to the community. All Trails is a great example of this. They’re free users who actually may update trail data and add new trails and things like that.

They’re directly contributing value to the business. So when we think about the hybrid monetization models, you need to be thinking about all these different buckets and how to derive value. How to extract value from the value you’re delivering. The free users of All Trails get a ton of value out of it.

But they’re also contributing back by seeing ads, by updating trails, by doing other things. So when you’re working on hybrid monetization, look at all the different ways that you’re giving away value, and then how you can receive some value back from those users. 

Shamanth: 

This evolution also mirrors what we’re seeing on the gaming side, where there’s been an increasing adoption of hybrid monetization, which is ads and IAP.

There are a handful of games that I’ve seen that have subscriptions, but that’s not proliferated nearly as much. So it’s interesting to see now that a lot of the low-hanging fruits are taken. A lot of subscription apps want to explore new and more interesting ways to monetize, which may not be just straight-up subscription plans.

David: 

B2B is an example of hybrid monetization. Calm, Headspace, and others, that’s a form of diversifying the income streams for this product that you’ve already created. There are so many different ways to accrue more value once you’ve created something that is valuable and B2B sales is just one other form of hybrid monetization.

Shamanth Rao: 

100 percent of a couple of wellness apps we work with, have B2B deals. One of them has live events. That could be another revenue stream just as well. 

And not to go too much aside, but in David’s episode on the Mobile UA Show, he talks about some of the early days of the app store. I would highly recommend it to anybody curious about it. What the app store was like from 2008 to 2010, and also the evaluation of subscription apps. 

Now, to switch gears a bit, let’s get into some of the nitty-grits of acquisition. We’ve talked about some of the missing data and some of the challenges with the acquisition on the mainstream channels.

SKAN 4.0 has been in the news. There’s been quite a bit of back and forth. Are you seeing adoption or success with Scan 4.0? Viren, if you want to take this. 

Virendra: 

No, unfortunately. We are aware that Facebook did roll that back out, but apart from that, we have not seen a lot of updates on that side. And apart from that also, I think across clients, So far, we have not seen a lot of adoption and we are sticking with SKAN 3.

Shamanth: 

I think we’ve been prepared for the longest time and it’s just not happened. John, you are very plugged in to all things SKAN, SKAN 4.0. What are you seeing? 

John: 

I’m seeing some interesting things, but let’s start with a bit of the bad news. So there’s a lot of hope out there. There’s little actual use. It’s just not out there enough. 

SKAN 4 is backward compatible with SKAN 3 so it costs you nothing to get ready, be ready. And it’s kind of easy to dip your toes in. Also, if you’re using a few networks that are almost all in on SKAN 4, there are some of those you can go for.

The Trade Desk is over 90% SKAN 4, Reddit is over 90% SKAN 4, Twitter’s around 80% SKAN 4 postbacks right now. Unity is close to that. Some other networks that are in the majority, let’s say, of their postbacks are SKAN for Jamf, SmatX, Personally, Mintegral, Datasheet, Applovin, and Liftoff.

So we know SKAN 4.0 is going to be better. You’re going to get more data over a longer period. We know crowd anonymity is better than privacy thresholds. It’s going to be easier to get into the data and get into the postbacks with payloads, and data payloads. It’s coming. Maybe it’s more Q2 to Q3 thing. 

Here’s the interesting part that I was going to get to. We saw that Meta said, okay, we’re going for SKAN 4 which was last year. We saw the bug that was deleting conversion values and they reverted and they’re still sub 5% SKAN 4 postbacks. And it’s like, they touched the fire, they touched the flame they got burnt and they pulled back and they’re like, screw this.

Also, it’s not in their best interests in a lot of ways to go SKAN 4, because they have their proprietary modeling technology. And you should always rely on Facebook measurement for your Facebook campaigns. Why would you use anything else? 

But what’s interesting is that I saw something that Google is doing and there’s been a trend with postbacks from Google, where it was sub 5% some months ago. Started creeping up, I saw 10%. Then recently I saw 15, 17, 18. I just checked this morning it’s currently at 20% over the last seven days. 20%t of the postbacks that Google is sending are SKAN 4. So what we might be seeing is a tipping point here.

We’re certainly seeing some incremental movement from Google if Google goes all in. The way Facebook went all in all of a sudden, that’s a big shift. That’s a huge shift and others have to jump into that. But many are low. And I’m just looking at it right now. A Moby, Snap, Iron Source Meta, all sub 5 percent SKAN 4 if Google says, Hey, this is working for us.

This is working for our customers. Our customers are demanding it, and they go all in and go to 50% or even 75% or something like that. All of a sudden, Meta will feel some pressure to come along, and the others will also. I think almost everybody is technically ready, and it’s just a question of flipping a switch.

Shamanth: 

And for those of you who aren’t yet ready, we have a book with our friends at Singular, the SKAN 4.0 Handbook that you guys can check out as well. We’ve been ready for a while, just that the real implementations never rolled around.

John, staying with you, the other thing that’s been on the horizon for a while is Google’s Privacy Sandbox for Android.

What’s happening with that? Is that on the horizon coming soon? What are you seeing?

John: 

You never get to the horizon. As soon as you walk closer to the horizon, the horizon moves. So here’s the deal with Privacy Sandbox. We were hearing and thinking that there was going to be a big announcement in the February, and March timeframe.

We’ve since heard that that’s been pushed back for at least a month or two. So that’s interesting. Our focus right now is making a big easy button because it’s not like we built it. We’ve integrated with Gameloft right now. Over the past few months and continuing. We’re testing attribution.

We built a protected audience integration. We just announced that literally, I want to say this week. And we’ll have some more tools coming out soon. So you can kind of role-play and test the Privacy Sandbox. If I did this, here’s what I’d get. The Privacy Sandbox isn’t SKAN. They’re approaching the question of privacy and advertising.

They’re both doing that, but Privacy Sandbox is a re-imagining of the entire ad tech stack. How ads are targeted, how you do retargeting, and remarketing, what ad somebody sees, and where those decisions happen. There’s a lot to do here. There’s a lot of work for the ad tech ecosystem to do here. And I do believe that we’ll see some interesting things this year, probably Q2, maybe as late as Q4, but it could slide as well.

There’s a lot to be done here. What I will say is wow’s a great time to test. So we saw that and we talked about that in the early days of SKAd Network when Apple rolled out. It was pre iOS 14.5 and we’re saying, β€œHey, you have it here. Test now.” Compare side by side with your IDFA campaigns. 

What can you learn? What can you see? How good can you get while you can check the cheat sheet? Right now is a good time to do that with GA ID. What that takes is some spare cycles, some spare people, some cash, some engineers, not everybody has that. But what we have seen is that the people who got good at SKAN early were able to use it, adopt it and be successful with much quicker than others.

Shamanth: 

it’s before everything is live that’s the best time to prepare. And it’s certainly something we’re doing with SKAN 4 and John, as you pointed out, that’s certainly something a lot of folks did with SKAN O\overall with ATT as well. 

A big aspect of, privacy changes has been measurement. That’s been one of the things that has thrown a wrench in the works. How are you guys looking at measurement? What are some of the challenges you’re seeing, Viren, if you want to take that? 

Virendra: 

I think there is no easy solution. We still do not know what the right approach is, but what we are trying to do over here is we are trying to be as diversified as we can while evaluating those metrics.

And then we have separate methodologies where we are taking blended metrics, where we are including organic data as well, so that directionally at least we are seeing either an increase or decrease. So we are much more informed in making our decisions. And then we are also having our campaigns analyzed via MMPs that we have.

We are doing revenue uplift analysis via App Store Connect. Then we have a proxy metric as ROAS analysis that we do via a peripheral. The data that we see on App Store Connect as well. So there are varied methodologies that we have employed, but to be frank, there is no easy solution. There are certain times when we are 100 percent not accurate about how we are doing in terms of the channels or the campaigns, but then again on top of that, we also have MediaMix modeling that we have heavily invested in.

We have built our internal resources. We have built on top of the open tool that we have from Facebook Robyn. We have been using that exclusively as well, just to analyze how our campaigns are doing, and how our channels are doing. And I think with that stack of software that we are using currently, directionally, we are somewhat suited to answer them.

But I would say that it’s still not 100%. It’s not an easy thing just to say which channel or which campaign is doing that much uplift with 100 percent accuracy. So still TBD and work in progress.

Shamanth: 

Or maybe it’s never going to get solved. And I think we just have to get the fact that we’re not going to get very clear deterministic attribution. And it’s going to be messy. You’re going to have to look at a bunch of sources to mix. So what is working and what is not? 

John, I know you tend to have a lot to say on the topic of measurement, and I’m curious what you think and what you’re seeing.

John: 

I loved it when I heard Viren say blended metrics because I think that is the way of the future.

Shamanth, you just said it. We’re going to have to get more used to uncertainty. We’ve talked about this in the past. Even when you had full deterministic IDFA and you have GA ID right now, just cause you have a last-click measurement doesn’t mean you know what happened. Doesn’t mean you know what caused an install, what caused the purchase, or what caused the subscription.

You got one little piece. You have one little window into consumer behavior, which is incredibly complex in a busy, noisy world. Blended metrics, however, are super interesting. So Singular calls are hybrid measurements. That involves a lot of first-party data. What’s happening in my app? I see that. I know that I can see my users.

I can see what they’re doing. I see what’s happening with them. That includes deterministic methodologies like SKAN that are limited and crippled in some ways for privacy and eventually privacy sandbox. That includes what you might get for IDFA if you got an ATT and whatever Google will do on the GA ID side eventually.

That includes the top funnel, what’s happening, impressions, clicks, all that click through those sorts of things. Even as those metrics get more muddled with people popping up app store views and everything like that every two seconds. And sometimes without anybody clicking or tapping as well.

And that also includes other deterministic as well as the MMM probabilistic, what might be happening. We’ll have something to announce there very, very shortly. Essentially knowing true organics. I think we’ve made a breakthrough and seeing all the various assists and being able to report a much cleaner number per channel and for organic is going to be interesting.

It’s very surprising when you get better at deducing what’s going on and seeing what each channel is contributing. People are making different buying decisions, different optimization decisions when they get that data as we’re testing this stuff. super interesting to see. Also, super interesting things coming on cohorts and longer cohorts.

Hybrid measurement I think is the future, and getting used to having to look at a bunch of different sources to know what’s going on is going to be the future. Now, the job of measurement companies is to take that complexity and simplify it.

And so there’s a lot of work happening there as well. 

David: 

One little interjection here is that there is one fully deterministic ad reporting left in the industry. You know what that is? Apple Search Ads. 

John: 

That may change. 

David: 

We’ll see, but for now, it’s a really good place to get some deterministic data. We’ve seen a lot of RevenueCat customers doing this cause we have an Apple Search Ads integration where you can get deterministic down to the keyword. How much people spent based on those specific keyword installs, and then that can actually feed back into the kind of ads that you’re putting on other networks, the kind of keywords that you’re putting in those ads and to find the keywords through ASA. 

And then find ways to work that into the rest of your marketing and ASO. Put that keyword in your title, do those kinds of things because it is the one place left where we do have deterministic attribution. 

Shamanth: 

I will nitpick and say it’s nearly deterministic because users who do opt out don’t get tracked, but it’s still miles ahead of everything else. Again getting to another aspect of acquisition, I know we talked briefly about creative earlier in this webinar. What are you seeing with subscription app creators that is,new, evolving and interesting? Katya, if you want to take that. 

Ekaterina: 

I think UGC is still definitely big. Whether UGC is produced with creators or UGC influencers and then reused for paid advertising which in my experience tends to work better, but it’s more like anecdotal evidence for now, even though it makes sense that it doesn’t because this space is somewhat familiar to users.

However I also see two sides of the spectrum these days when we do competition research or when I also browse my own feed on social media. There are this very well thought through ads. And you can see how a lot of money went into their production with great storytelling with a lot of humor.

For example, Headspace has amazing ads recently which I’ve seen and on the other hand, you also can see a lot of static ads with a lot of text on them. So this 20 percent text limit on Facebook ads is really gone. Nobody cares about that anymore. And still, you can see how If you check roughly how long these ads have been running, both approaches work.

So on the one hand, there are these ads where you have to really come up with a very compelling narrative and great visuals. And on the other hand, something very different, which stands out, and makes you zoom in on the thumbnail, this also works. 

So for me, these are two interesting trends that are evolving and I also wanted to say that I think it’s great with all the recent changes and into privacy to some extent as well that we can actually now super easily monitor the competition and we can easily see who is running what.

And who is testing what and what might be working for whom with the transparencies, the transparency center with, meta ad library TikTok library. It’s become super easy and ton of inspiration there. 

Shamanth: 

Interesting. Viren, What are you seeing with creatives lately? 

Virendra: 

With respect to AI and creators, I think what Ekaterina said, UGC definitely remains the biggest lever to drive performance across channels.

But in terms of AI, I think what has helped us to develop and conceptualize those UGCs has become more efficient and less time-intensive. I mean a couple of custom GPTs that you only created for month script image maker, which has helped our organization and the other organization as well, whenever we speak.

So now it’s not a matter of days to come up with new concepts, to do the research, to find some inspiration. UGC scripts, display creatives, video creatives, everything can be done within minutes now and not hours. From conceptualizing those creatives to doing some research and then the production as well.

For example, we have HeyGen, Dall-E, Genmo, and Replicate among other tools. Those tools have helped us and effectively what it has been doing is we can test out a higher volume of creatives now.

We have so many variants that we can develop with AI, so many display creatives. In terms of testing out those creatives, the testing process has become a lot faster. We are identifying winning elements out of those, we can iterate very quickly. And that has eventually helped us to drive better performance.

AI has helped us much more on the creative side than some of the other offerings it has been doing lately.

Shamanth: 

100%. And I think AI has changed the game a lot. And for those that weren’t familiar with the details that Viren mentioned, we’ve made a couple of custom GPTs that are completely free and we send it out in our emails.

We’ve trained a custom GPT on UGC scripts. We’ve trained a custom GPT on our top-performing display ads. With these, things become a lot easier, even in the ideation and conceptualization and just building on what is winning. So I think what we’re seeing work is really using AI tools to build on your own proprietary learnings.

And I think that is huge for any performance marketing team lately. That brings us very close to the end 

We just have a handful of minutes left. Staying on the theme of generative AI tools. What are you guys seeing around AI tools that are interesting for subscription apps, John, if you want to take that?

John: 

I’m super interested. I’m deep in it. I loved hearing what Virendra was talking about what he’s using. And you’ve built some custom GPTs. That’s news to me. That’s very cool. It’s awesome. I’m looking forward to what the platforms are going to do with generative AI and ads that are fully tailored to me because Meta knows me or Google knows me because I’m a logged-in user and they have a lot of history and data on me.

That’s going to be tougher for the independent ad networks to do because they don’t have their own walled garden. But I think there’s going to be amazing things there. It’s almost like cheating on what you’re doing Shamanth and Virendra but I like it. 

Shamanth: 

We’ll share some of the custom GPTs in the email that will go out. They’re all free. They’re public in the GPT store. 

We are very close to time. So I want to take a moment to take the one audience question that seems unanswered. How is aggressive marketing techniques from Demo and Shane impacting the space and how are our budgets positioned iare n 2024? Is there anyone on the panel that wants to take that?

John: 

I can speak generally about it, but obviously if they’re pouring tons of money into retail, it’s going to drive up costs. And it’s going to take airspace from other apps, and that has a follow-on effect because you see an organic multiplier as they start to continually hit the tops of the charts and people wonder, why is it up there?

I guess I should try it. I’ve been tempted myself, except I have too much stuff already. Why do I want more stuff? But it’s tougher for other apps to compete. It’s like the TikTok playbook. TikTok got big with massive ad campaigns, and billions of dollars spent over a few years. There were a lot of other factors.

The app was amazing. The experience was incredible. It was completely addictive. So you had one and you had one and it added up to three. It’s tough to compete with. 

Shamanth: 

Indeed. We’ll see. I’m sure there’s a ceiling to how much they can spend. And we’ll see how that lasts. They have to make the LTV back too. And we’ll have to see what that ceiling is.

But well we are at the time. And this was an amazing webinar. I certainly learned a ton. I’m going to go back to the notes from this. Thank you so much to our amazing panelists so open with their sharing. Thank you to our amazing audience. And for those of you who couldn’t make it, you’ll catch this on the mobile UA show podcast and our email newsletter as well.

WANT TO SCALE PROFITABLY IN A POST IDENTIFIER WORLD?

Get our free newsletter. The Mobile User Acquisition Show is a show by practitioners, for practitioners, featuring insights from the bleeding-edge of growth. Our guests are some of the smartest folks we know that are on the hardest problems in growth.